Skip to content

Y—Analytics: a generic analytics wrapper you can put around most any analytics framework

License

Notifications You must be signed in to change notification settings

codeandtheory/yanalytics-ios

Repository files navigation

Y—Analytics
An injectable analytics wrapper for Apple platforms.

This lightweight framework provides a generic wrapper for analytics SDK's that is suitable for Dependency Injection. Its objective is to avoid tightly coupling an application with a specific analytics provider.

Licensing

Y—Analytics is licensed under the Apache 2.0 license.

Documentation

Documentation is automatically generated from source code comments and rendered as a static website hosted via GitHub Pages at: https://yml-org.github.io/yanalytics-ios/

Usage

AnalyticsEngine

is an abstraction meant to represent any analytics package, whether that be Firebase, Adobe, any other 3rd party analytics vendor, or your own custom implementation.

The point is to abtract away the specifics of choice of analytics vendor, so that it may be swapped at any time without incurring major tech debt.

It also encourages the use of dependency injection to provide the analytics engine to components that require it. This should make it easier to write unit tests for all these components (by either injecting a mock or even nothing at all).

AnalyticsEngine has exactly one method:

public protocol AnalyticsEngine {
    /// Track an analytics event
    /// - Parameter event: the event to log
    func track(event: AnalyticsEvent)
}

AnalyticsEvent

is an enum with a few cases for typical events.

public enum AnalyticsEvent {
    /// Screen view event
    /// - Parameter screenName: screen name
    case screenView(screenName: String)
    /// User property event
    /// - Parameters:
    ///   - name: user property name
    ///   - value: user property value
    case userProperty(name: String, value: String)
    /// Generic event
    /// - Parameters:
    ///   - name: event name
    ///   - parameters: event metadata to track
    case event(name: String, parameters: Metadata?)
}

CompoundAnalyticsEngine

is a helper implementation of AnalyticsEngine that lets you specify an array of engines. This would allow you to combine a logger together with your actual analytics engine (e.g. Firebase) to assist with debugging.

ConsoleAnalyticsEngine

is a simple implementation of AnalyticsEngine that logs output using print statements. (For most cases you should use LoggerAnalyticsEngine instead.)

LoggerAnalyticsEngine

is a simple implementation of AnalyticsEngine that logs output using Apple's' Logger api.

Installation

You can add Y—Analytics to an Xcode project by adding it as a package dependency.

  1. From the File menu, select Add Packages...
  2. Enter "https://github.com/yml-org/yanalytics-ios" into the package repository URL text field
  3. Click Add Package

Contributing to Y—Analytics

Requirements

SwiftLint (linter)

brew install swiftlint

Jazzy (documentation)

sudo gem install jazzy

Setup

Clone the repo and open Package.swift in Xcode.

Versioning strategy

We utilize semantic versioning.

{major}.{minor}.{patch}

e.g.

1.0.5

Branching strategy

We utilize a simplified branching strategy for our frameworks.

  • main (and development) branch is main
  • both feature (and bugfix) branches branch off of main
  • feature (and bugfix) branches are merged back into main as they are completed and approved.
  • main gets tagged with an updated version # for each release

Branch naming conventions:

feature/{ticket-number}-{short-description}
bugfix/{ticket-number}-{short-description}

e.g.

feature/CM-44-button
bugfix/CM-236-textview-color

Pull Requests

Prior to submitting a pull request you should:

  1. Compile and ensure there are no warnings and no errors.
  2. Run all unit tests and confirm that everything passes.
  3. Check unit test coverage and confirm that all new / modified code is fully covered.
  4. Run swiftlint from the command line and confirm that there are no violations.
  5. Run jazzy from the command line and confirm that you have 100% documentation coverage.
  6. Consider using git rebase -i HEAD~{commit-count} to squash your last {commit-count} commits together into functional chunks.
  7. If HEAD of the parent branch (typically main) has been updated since you created your branch, use git rebase main to rebase your branch.
    • Never merge the parent branch into your branch.
    • Always rebase your branch off of the parent branch.

When submitting a pull request:

  • Use the provided pull request template and populate the Introduction, Purpose, and Scope fields at a minimum.
  • If you're submitting before and after screenshots, movies, or GIF's, enter them in a two-column table so that they can be viewed side-by-side.

When merging a pull request:

  • Make sure the branch is rebased (not merged) off of the latest HEAD from the parent branch. This keeps our git history easy to read and understand.
  • Make sure the branch is deleted upon merge (should be automatic).

Releasing new versions

  • Tag the corresponding commit with the new version (e.g. 1.0.5)
  • Push the local tag to remote

Generating Documentation (via Jazzy)

You can generate your own local set of documentation directly from the source code using the following command from Terminal:

jazzy

This generates a set of documentation under /docs. The default configuration is set in the default config file .jazzy.yaml file.

To view additional documentation options type:

jazzy --help

A GitHub Action automatically runs each time a commit is pushed to main that runs Jazzy to generate the documentation for our GitHub page at: https://yml-org.github.io/yanalytics-ios/

About

Y—Analytics: a generic analytics wrapper you can put around most any analytics framework

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages